Is there evidence that social class at birth increases risk of psychosis? A systematic review

Abstract

Background:

In the 1950s, researchers showed an association between low socio-economic status (SES) and psychosis. Two competing theories social causation and social drift were proposed to explain the findings. In the intervening years, contrasting evidence emerged as some studies showed no association between SES and schizophrenia. At present, the nature of the relationship is still unclear; currently, there are no reviews in the literature examining the association between social class at birth and psychosis.

Aim:

To search the literature to clarify the relationship between social class at birth, measured by paternal occupation at birth, and the risk of adult-onset psychosis.

Methods:

A systematic search of the literature using a combination of keywords in Group 1 together with the keywords in Group 2 was performed in October 2012 in the following online databases: (a) MEDLINE (1946–2012), (b) PubMed, (c) Embase (1980–2012), (d) PsycINFO (1806–2012) and (e) Web of Science (1899–2012). Reference lists were also hand searched. The search provided 3,240 studies; following screening of the titles and abstracts by inclusion and exclusion criteria and quality assessment of the full text, 14 studies were identified to be appropriate for the review. The keywords used for the search were as follows: Group 1 – social class, social status, socioeconomic, socio-economic, SES; Group 2 – psychosis, psychoses, schizophrenia.

Results:

Seven studies showed an association between low SES and psychosis. Four studies showed no association, and three studies showed an association with high SES.

Conclusion:

There is not enough evidence to support the association between social class and psychosis. While some findings showed an association between low social class and psychosis, there were a number of conflicting studies showing no association or a link with higher social class. Interestingly, the results followed a temporal pattern, as all the studies conducted after 2001 supported an association between low SES at birth and psychosis. Four of the six studies employed a prospective design with large sample populations, indicating the need for further investigation.

Keywords

Social class socio-economic status psychosis schizophrenia

Introduction

Socio-economic status and psychosis

An individual’s rank in society, determined by the rewards and resources he or she earns is known as his or her socio-economic status (SES). SES is primarily indexed by measures of income, occupation and educational attainment (Liberatos, Link, & Kelsey, 1988). In the literature, the terms social class and SES are used interchangeably, and in this article, they will be viewed as synonyms (Eaton & Harrison, 2001). An individual’s social class at birth is predominantly determined by measuring parental social class at birth, which is commonly defined by parental occupation on a 6-point scale (Table 1). Historically, paternal occupation is more frequently used to define individual SES at birth compared to maternal occupation (Byrne et al., 2004; Werner, Malaspina, & Rabinowitz, 2007).

Table 1.

The six categories of social class (Central Statistics Office, 1986).

Social class scale
Social class I – higher professional, higher managerial, proprietors and farmers with ≥200 acres (81 hectares)
Social class II – lower professional, lower managerial and farmers with 100–199 acres
Social class III – other non-manual and farmers with 50–99 acres
Social class IV – skilled manual and farmers with 30–49 acres
Social class V – semi-skilled manual and farmers with <30 acres
Social class VI – unskilled manual

In the mid-1950s, research in the United States and England showed a positive association between low SES and psychosis (Hollingshead & Redlich, 1958). For example, Brooke (1959) analysed first admissions for schizophrenia in England between 1949 and 1952. The results showed that for single men, the rate in social class V was four times higher than social class I. Individuals in a lower social class were also more likely to be compulsorily admitted into hospital and have a worse clinical prognosis (Cooper, 1961). Association between low SES and psychotic illness was also shown when using other measures of low SES such as educational attainment, area of residence and ethnicity (Byrne, Agerbo, Eaton, & Mortensen, 2004; Castle, Scott, Wessely, & Murray, 1993; Eaton, 1985; Harrison, Gunnell, Glazebrook, Page, & Kwiecinski, 2001). Two competing theories, social causation and social drift, were proposed to explain these findings. The social causation hypothesis suggests that there is a high prevalence of patients with psychosis in the lower social classes because low SES is causative of the condition.

On the contrary, the social drift hypothesis states that low SES among patients with psychosis is a consequence of a drift down in the society due to the illness. For example, Dohrenwend et al. (1992) suggest that these individuals were not disadvantaged from birth but were handicapped because of their condition leading to reduced educational attainment and occupational progression, resulting in a subsequent shift down the social hierarchy. Evidence supporting this hypothesis shows a discrepancy between paternal SES at birth and SES of the affected individual (Goldberg and Morrison, 1963; Mäkikyrö et al., 1997; Wiersma, Giel, De Jong, & Slooff, 1983). Goldberg and Morrison (1963) compared paternal SES of individuals aged 25–34 years first presenting with schizophrenia with paternal SES of individuals from the general population. Although there was a high prevalence of individuals in social class V, paternal SES did not significantly differ to the general population, suggesting that low SES is not causally associated with psychosis. Further examination of the case group showed a pattern of academic and work achievement decline beginning in adolescence. These results led many investigators to conclude that ‘social drift’ was sufficient in explaining the epidemiological associations.

In the intervening years, contrasting evidence has emerged to show that there is no association between individual SES at birth and adult-onset schizophrenia (Hare, Price, & Slater, 1972; Jones, Rodgers, Murray, & Marmot, 1994; Wender, Rosenthal, Kety, Schulsinger, & Welner, 1973). At present, the nature of the relationship is unclear as there are conflicting study results in the literature. There are strong parallels between the indigenous populations investigated in previous studies and the inner city immigrants of England, in terms of low paternal SES at birth and elevated rates of psychotic illnesses (Fearon et al., 2006). Indeed, a Swedish cohort study by Hjern et al. (2004) showed that the risk ratio (RR) for all ethnic minorities could be eliminated after adjusting for socio-economic differences. In contrast, Kirkbride, Barker, Yang, Jones, and Coid (2008) showed that Black and ethnic minority groups had elevated rates of psychotic disorders after adjustment for age, gender and individual SES. Thus, there have been strong calls to reappraise evidence examining the relationship between social class and psychosis using a systematic approach.

Aim of this systematic review

This systematic review aims to clarify whether a low social class at birth is associated with adult-onset psychosis. Since many studies report contrasting results dependent on criteria used to define SES, this review will focus on studies using paternal social class, defined by occupation at the time of birth. Results from this study will allow verification of the epidemiological associations underpinning the social causation and drift hypotheses, and therefore are essential for their endorsement for further investigation. Identification of environmental risk factors for schizophrenia is important in furthering our understanding of the condition and may give insight for its prevention.

Methods

Search strategy

A search strategy was used to identify published studies investigating the relationship between individual SES at birth and adult-onset psychotic illnesses. In this review, individual SES is defined by paternal SES at birth. A search on October 2012 in five online databases – (a) MEDLINE (1946–2012), (b) PubMed, (c) Embase (1980–2012), (d) PsycINFO (1806–2012) and (e) Web of Science (1899–2012) – was performed. Keywords in Group 1 and Group 2 were combined as free text words in the title and abstract. Truncated and wildcard symbols were not used as the search results were too broad in preliminary searches:

Group 1 – social class, social status, socioeconomic, socio-economic, SES

Group 2 – psychosis, psychoses, schizophrenia

The search identified 3,240 studies across the five databases, of which 1,664 studies were originals. Additionally, 12 studies were identified from hand searching of reference lists, to capture studies not found in the database searches. This provided 1,676 studies which were screened via the title and abstract according to the specified criteria below. If it was unclear whether the study was suitable, the full text was retrieved. For completeness, the grey literature was also searched and returned no relevant studies.

Inclusion and exclusion criteria

Studies were selected for further evaluation if they (a) were published in English in a peer-reviewed journal, (b) described empirical research methods, (c) included adults aged 18–64 years diagnosed with psychotic symptoms or schizophrenia and (d) used paternal occupation at birth as one measure to define individual SES. Reviews, abstracts and reports were excluded. Studies describing the relationship between social class and severity of psychotic symptoms were also selected out, as this was not deemed to reflect association between the variables.

Methodological quality assessment

After screening of inclusion and exclusion criteria, 14 studies remaining underwent methodological quality assessment (Table 2) by the author and an independent researcher. Before the studies were assessed, eight desirable criteria were identified through discussion, by which the studies were rated against independently. One point was awarded for each criterion met. Studies were awarded a rating of ‘high’ for scoring 6–8 points, ‘medium’ for 3–5 points and ‘low’ for 0–2 points. If there was a disagreement in the quality score, a discussion was held to reach a consensus.

Table 2.

Methodological quality assessment criteria.

Subject criteria
Were the eligibility criteria identified?
Were there sufficient attempts to avoid a selection bias through appropriate randomisation of the sample population?
Was a validated tool used to diagnose individuals (i.e. using ICD or DSM criteria for psychoses or schizophrenia)?
Was there an adequate description of how social class is defined?
Outcome measurement and reporting
Were the demographics of individuals clearly described? (at least two of the following) • Age • Sex • Individuals occupation
Was there a clear description of how paternal SES is defined?
Were the responses of individuals appropriately anonymised and free from author’s influence?
Were appropriate methods of data reporting used (i.e. mean data with standard deviation, dichotomous outcomes, median with range or sufficient data to enable extraction of any of these)?

ICD: International Classification of Diseases; DSM: Diagnostic and Statistical Manual of Mental Disorders.

Search yield

None of the articles meeting the inclusion and exclusion criteria were excluded from the review following methodological quality assessment (Table 3).

Table 3.

Methodological characteristics of reviewed studies.

Author	Design	Study location	Source of participants	Sample size (n)	Individual demographics	Validated tool for diagnosis	Primary condition assessed	Appropriate method for data reporting	Central findings	Methodological quality assessment rating
Goldberg et al. (1963)	Retrospective	England	Hospital	369	No	–	Adult-onset schizophrenia	Yes	Paternal SES similar to general population	4
Hare, Price, and Slater (1972)	Retrospective	England	Hospital	624	No	–	Adult-onset schizophrenia	Yes	Paternal SES similar to general population	6
Wender, Rosenthal, Kety, Schulsinger, and Welner (1973)	Retrospective	Denmark	Community	5,483	No	–	Adult-onset schizophrenia	Yes	Paternal SES similar to general population	5
Castle et al. (1993)	Retrospective	United Kingdom	Community	256	No	Research diagnostic criteria	Adult-onset schizophrenia	Yes	Low paternal SES increased risk for schizophrenia	7
Jones, Rodgers, Murray, and Marmot (1994)	Longitudinal	United Kingdom	Community	5,362	No	DSM-III-R	Adult-onset schizophrenia	Yes	No association between low paternal SES and schizophrenia	5
Mäkikyrö et al. (1997)	Longitudinal	Finland	Community	11,017	Yes	DSM-III-R	Adult-onset schizophrenia	Yes	High paternal SES increased risk for schizophrenia	6
Timms (1998)	Longitudinal	Sweden	Community	15,117	No	ICD-8	Adult-onset schizophrenia	Yes	High paternal SES increased risk for schizophrenia	5
Mulvany et al. (2001)	Retrospective	Ireland	Community	704	No	ICD-9	Adult-onset schizophrenia	Yes	High paternal SES increased risk for schizophrenia	6
Harrison, Gunnell, Glazebrook, Page, and Kwiecinski (2001)	Retrospective	England	Community	280	No	ICD-9	First episode psychosis	Yes	Low paternal SES increased risk for schizophrenia	4
Byrne, Agerbo, Eaton, and Mortensen (2004)	Retrospective	Denmark	Community	200,294	Yes	ICD-10	Adult-onset schizophrenia	Yes	Low paternal SES increased risk for schizophrenia	7
Wicks, Hjern, Gunnell, Lewis, and Dalman (2005)	Longitudinal	Sweden	Community	2,100,000	No	ICD-10	Adult-onset schizophrenia	Yes	Low paternal SES increased risk for schizophrenia	5
Werner, Malaspina, and Rabinowitz (2007)	Longitudinal	Israel	Community	71,165	No	ICD-9	Adult-onset schizophrenia	Yes	Low Paternal SES increased risk for schizophrenia	5
Corcoran, Perrin, Harlap, and Deutsch (2009)	Longitudinal	Jerusalem	Community	88,829	Yes	ICD-10	Adult-onset schizophrenia	Yes	Low Paternal SES increased risk for schizophrenia	7
Goldberg et al. (2011)	Longitudinal	Israel	Community	811,487	No	ICD-10	Adult-onset schizophrenia	Yes	Low Paternal SES increased risk for schizophrenia	5

SES: socio-economic status; ICD: International Classification of Diseases: DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders–Third Edition, Revised.

After the selection process, 14 articles were identified as suitable for the review (Figure 1).

Figure 1.

Search strategy of this systematic review.

Results

From the 14 studies included in this review, 7 showed that a low paternal SES is associated with an increased risk for schizophrenia (Byrne et al., 2004; Castle et al., 1993; Corcoran, Perrin, Harlap, & Deutsch, 2009; Goldberg et al., 2011; Harrison et al., 2001; Werner et al., 2007; Wicks, Hjern, Gunnell, Lewis, & Dalman, 2005). One longitudinal study showed no association between low paternal SES and risk for schizophrenia (Jones et al., 1994). Three studies showed that an increased paternal SES is associated with schizophrenia (Mäkikyrö et al., 1997; Mulvany et al., 2001; Timms, 1998). Finally, three retrospective studies showed that paternal SES of people with schizophrenia is similar to the general population (Goldberg and Morrison, 1963; Hare et al., 1972; Wender et al., 1973). Interestingly, results from the included studies appeared to follow a temporal pattern, as all of the studies conducted since 2001 supported an association between low paternal SES and psychosis (Byrne et al., 2004; Corcoran et al., 2009; Goldberg et al., 2011; Harrison et al., 2001; Werner et al., 2007; Wicks et al., 2005).

Paternal SES is associated with psychotic illness

A longitudinal study by Goldberg et al. (2011) showed that in a large sample population of Israeli adults (n = 811,487), individuals with a low paternal SES are at a higher risk of hospitalisation for schizophrenia compared to high paternal SES individuals (hazard ratio (HR) = 1.193, 95% confidence interval (CI) = 1.091–1.303). There was a follow-up of 1–15 years between first assessment of individual SES and hospitalisation for schizophrenia. A similar association was found by Werner et al. (2007) in a longitudinal study of Israeli adults with a 13-year follow-up (n = 71,165). The risk was present when measuring either paternal occupational prestige (low vs high, odds ratio (OR) = 1.39, 95% CI = 1.10–1.78, p < .01) or maternal occupational prestige (low vs high, OR = 1.99, CI = 1.29–3.05, p < .001). Similarly, a case-control study of British adults (n = 280) by Harrison et al. (2001) showed an increase in risk for individuals in social class IV/V at birth compared to social class I/II (OR = 2.6, CI = 1.0–6.9, p = .010). This increase in risk persisted after adjusting for advanced parental age, sex, duration of marriage and birth order. Castle et al. (1993) echoed the previous findings in another small case-control study (n = 256). The authors showed that the risk of British individuals with fathers in ‘manual’ occupations was twice as high compared to ‘non-manual’ occupations (OR = 2.1, 95% CI = 1.2–3.7).

Although the previous studies show an increase in risk for the low social classes, there is weak evidence that the risk follows a social-class gradient. Goldberg et al. (2011) found a significant paternal social-class gradient for the risk of hospitalisation for schizophrenia. As paternal SES increased, there was a corresponding slight decline in risk (χ² = 53.925, df = 19, p < .001). In contrast, two retrospective studies showed no gradient of risk for schizophrenia. First, Corcoran et al. (2009) used the highest paternal social class in a 6-point scale to compare with the other five ranks and found no significant ascending or descending trend (p > .05). Second, a case-control study (n = 352) of Irish adults showed no consistent trend in OR across the six paternal social classes (Mulvany et al., 2001). Furthermore, Corcoran et al. (2009) suggest that the association with psychotic illness is only present in the lowest paternal social class. A sample population (n = 88,829) born between 1964 and 1976 were followed up between 22 and 34 years. Only individuals in the lowest social class at birth had a moderate increase in risk for schizophrenia regardless of the first four social classes used for comparison (RR = 1.4; 95% CI = 1.1–1.8, p = .002).

Byrne et al. (2004) analysed a large case-control study (n = 200,294) of Danish adults. First, the researchers assessed the contribution of general background factors to risk of schizophrenia and found that an increased risk was associated with co-morbid psychiatric disorders, family history of schizophrenia and a history of attempted suicide. After controlling for these background factors, they found that an increased risk for schizophrenia was associated with paternal employment for less than 100% of the year (incidence rate ratio (IRR) = 1.21, 95% CI = 1.11–1.33) when comparing other social classes to blue-collar workers (reference group). Interestingly, increased risk was also associated with maternal employment for less than 100% of the year (IRR = 1.21, 95% CI = 1.11–1.31) and being outside the workforce (IRR = 1.29, 95% CI = 1.17–1.42). Similar results were also reported by Wicks et al. (2005) in a longitudinal study using nationally linked data of Swedish adults (n = 2,100,000) between 1970 and 1990. Individuals were followed up between 1987 and 2002, and the results showed that the HR for people with schizophrenia with unemployed parents in a single-parent household was 2.0 (95% CI = 1.9–2.2). The HR for individuals with other psychotic illnesses was 1.8 (95% CI = 1.7–2.0).

In contrast to the previous studies, a longitudinal study by Mäkikyrö et al. (1997) found an association between high paternal SES at birth and adult-onset psychotic illness. A sample from the Northern Finland 1966 birth cohort (n = 11,017) was studied between the ages 16 and 27 years. The authors found that the incidence of schizophrenia was significantly higher (p < .05) in individuals of social class I (1.14%; 9/792) compared to social classes II–V (0.47%;48/10,225) at the age of 23 years. After this age, no new cases of psychotic illness was reported in social class I. Two other studies showed similar results. First, Wender et al. (1973) in an adoption study showed that there is a significant positive correlation (p < .05) between high adoptive paternal SES and risk for psychotic illness in offspring. Second, the OR was 0.49 (95% CI = 0.40–0.85, p < .0029) for people in low social classes compared to high social class in a small case-control study (n = 352; Mulvany et al., 2001), suggesting a slightly elevated risk for individuals in high social class. Furthermore, Timms (1998) found no association between low social class and psychotic illness in a prospective longitudinal study (n = 15,117). Curiously, individuals in the middle social classes had a non-significant increase in relative risk (RR = 1.60 95% CI = 0.95–2.72, p > .05) compared to the bottom social classes.

Paternal SES is not associated with psychotic illness

A pilot study by Goldberg et al. (1963) showed that in a small national sample (n = 369) of individuals with schizophrenia aged 25–34 years, there was an elevated number of observed patients in social class V than expected (observed = 90, expected = 34). Further comparison between these individuals and the general population aged 20–44 years showed very similar distributions of paternal social class, and thus suggests no association with psychotic illness. Similar results were reported by Hare et al. (1972) in a case-control study (n = 624) using a sample of British adults admitted into Bethlam-Maudsley hospital. The authors found that there was no significant difference between social classes of psychotic individuals and the general population.

Wender et al. (1973) used an interesting study design to investigate the relationship between social class and risk of psychosis. The authors studied a sample of adopted individuals (n = 5,483) using a case-control design. The sample was drawn from adoption lists between 1924 and 1947 in Copenhagen, and paternal SES information was obtained for both biological and adoptee parents. The sample population was then compared with the national psychiatric register in 1964. The authors showed there was no difference between the general population paternal SES and paternal SES of biological fathers (χ² = 1.86, df = 2, p > .05) or adoptee fathers (χ² = 0.75, df = 2, p > .05). Finally, a longitudinal study by Jones et al. (1994) analysed a British 1946 birth cohort (n = 5,362), with individuals followed up to age 43 years (1989). The results did not provide evidence that there is a significant association between low paternal social class and adult-onset schizophrenia (p > .05).

Discussion

The main strength of this study, to the author’s knowledge, is that it is the first systematic review to investigate the association between paternal SES at birth and risk of adult-onset psychosis. The results were variable; half the studies supported an association between low paternal SES and psychotic illness, and the other half showed either no association or association with high paternal social class. Four of the most recent studies supporting an association with low paternal SES at birth (Corcoran et al., 2009; Goldberg et al., 2011; Werner et al., 2007; Wicks et al., 2005) used a prospective longitudinal design with large sample populations (71,165–2,100,000) to reach the same conclusion, and thus possessed high validity. The three studies which showed no association were all undertaken between 1963 and 1973 (Goldberg et al., 1963; Hare et al., 1972; Wender et al., 1973). These studies were retrospective and used comparatively low sample populations (369–5,483), suggesting a low validity and weak statistical power. In contrast, three studies (Mäkikyrö et al., 1997; Mulvany et al., 2001; Timms, 1998) that supported an association with high SES at birth were performed between 1997 and 2001. Two studies had high validity and statistical power as large sample populations (11,017–15,117) were analysed in a longitudinal design (Mäkikyrö et al., 1997; Timms, 1998).

After a review of the general trends, it cannot be concluded that there is an association between social class at birth and psychotic illness in later life due to variability of the individual study results. Further investigation into the temporal aspects of the findings may allow further insight into this relationship, as all of the studies conducted after 2001 supported the association between low SES and psychosis, four of which are of gold standard longitudinal design with large sample populations. If the total number of cases were to be tabulated from the 14 studies included in the review, the vast majority of cases would come from studies which support the relationship between low paternal SES and increased risk for psychosis in later life. From these results, it can be hypothesised that future observational studies will also support this association, but a cautious approach should be taken, as unfortunately, it was not possible to conduct a meta-analysis of the data due to study heterogeneity.

The social causation theory suggests adversity and chronic strains associated with a low SES are causal factors of psychosis (Werner et al., 2007). First, parents of low SES may have fewer resources for their children leading to harsher residential environments, fewer educational opportunities and barriers to social networks leading to social isolation. Second, a body of evidence suggests that some patients with psychosis have the condition as a consequence of neurodevelopment deviance, where an insult to the developing brain leads to symptoms manifesting later on in life (Weinberger, 1987). Limited financial assets in low-SES households (Brown, Susser, Jandorf, & Bromet, 2000) may lead to insufficient prenatal and postnatal care, potentially causing brain insult due to obstetric complications, malnutrition and a low birth weight (Castle et al., 1993; Koenig, Kirkpatrick, & Lee, 2002). This causal relationship appears to only affect the lowest social classes. For example, both Harrison et al. (2001) and Corcoran et al. (2009) found the risk for schizophrenia was significant in only the bottom two social classes. Furthermore, there was no evidence for gradient of risk for schizophrenia (Mulvany et al., 2001). The associations between high SES at birth and psychosis may reflect underlying relations between education and social class (Mäkikyrö et al., 1997; Wender et al., 1973). Individuals of high educational attainment are more likely to be in a higher social class. Both Wiersma et al. (1983) and Byrne et al. (2004) have shown an increased risk of schizophrenia for individuals with high educational attainment. It has been hypothesised that the genetic predisposition for a high IQ may be associated with the genetic basis of schizophrenia (Aylward, Walker, & Bettes, 1984).

Alternatively, a family history of psychiatric illness is a potential mechanism by which an individual’s risk for schizophrenia increases during later life (Werner et al., 2007). The parents may also suffer from psychosis and can potentially transmit a genetic risk to their offspring. Patients with schizophrenia experience widespread cognitive deficits which can severely impair social and vocational functioning (Heinrichs & Zakzanis, 1998), and there is evidence to suggest this cognitive decline can present before the onset of psychotic symptoms (Cosway et al., 2000). As the parent drifts into a lower SES due to their condition and associated social disabilities, the child is thus born into a lower social class (Mortensen et al., 1999).

Future horizons

There are four main limitations of this study. First, there may be publication bias as only published studies were included in this review. Relevant studies may not have been found using the database and reference searches. Second, paternal psychopathology was not measured, and thus, confounding genetic effects could not be ruled out. Third, study heterogeneity made it difficult to compare the studies. This is exacerbated by the limited number of studies in the literature. Fourth, only paternal occupation was used to define SES in this review and so may not be truly representative of social class. To overcome these limitations, a future review should analyse studies using large sample populations with a prospective longitudinal study design. Ideally, multiple markers of social class should be present together with a measure of parental psychopathology.

Interestingly, Byrne et al. (2004) showed that after adjusting for general background and paternal SES, individuals with low SES in adulthood were at a higher risk of psychosis. This risk was associated with low levels of employment (Crude RR = 9.63, 95% CI = 8.86–10.46), wealth (Crude RR = 2.71, 95% CI = 2.53–2.91) and income (Crude RR = 20.77, 95% CI = 18.72–23.05). Additionally, individuals born in inner city London were at higher risk (OR = 2.1, 95% CI = 1.2–3.7) compared to being born elsewhere (Castle et al., 1993). Furthermore, after controlling for paternal SES, sex, year of birth and fathers’ age at birth, individuals in a low-SES area were still at increased risk compared to high-SES areas (OR = 1.26, 95% CI = 1.05–1.52, p < .05; Byrne et al., 2004). In light of these epidemiological findings, a review investigating association between individual SES and psychosis should be considered, as at present, there are no systematic reviews in the literature. At present, it cannot be concluded that paternal SES at birth is associated with psychosis.

Footnotes

Acknowledgements

The author thanks Robin Murray and Simona Stilo (Institute of Psychiatry, London) and the anonymous reviewers for their comments and suggestions on this manuscript. The author also thanks Siobhan Lynch and Simon Wessely (Institute of Psychiatry, London) for their continued support.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Aylward

Walker

Bettes

(1984). Intelligence in schizophrenia: Meta-analysis of the research. Schizophrenia Bulletin, 10, 430–459.

Brooke

E. M.

(1959). A national study of schizophrenic patients in relation to occupation. In: Second International Congress for Psychiatry, Zurich 1957: Congress Report. Vol 3. Zurich: Orelli Fussli, pp. 52–63.

Brown

A. S.

Susser

E. S.

Jandorf

Bromet

E. J.

(2000). Social class of origin and cardinal symptoms of schizophrenic disorders over the early illness course. Social Psychiatry and Psychiatric Epidemiology, 35, 53–60.

Byrne

Agerbo

Eaton

Mortensen

P. B.

(2004). Parental socio-economic status and risk of first admission with schizophrenia: A Danish national register based study. Social Psychiatry and Psychiatric Epidemiology, 39, 87–96.

Castle

D. J.

Scott

Wessely

Murray

R. M.

(1993). Does social deprivation during gestation and early life predispose to later schizophrenia? Social Psychiatry and Psychiatric Epidemiology, 28, 1–4.

Central Statistics Office. (1986). Census of population 1986: Classification of occupations. Dublin, Ireland: Stationery Office.

Cooper

(1961). Social class and prognosis in schizophrenia. British Journal of Preventive & Social Medicine, 15, 17–41.

Corcoran

Perrin

Harlap

Deutsch

(2009). Effect of socioeconomic status and parents’ education at birth on risk of schizophrenia in offspring. Social Psychiatry and Psychiatric Epidemiology, 44, 265–271.

Cosway

Byrne

Clafferty

Hodges

Grant

Abukmeil

… Lawrie

S. M.

(2000). Neuropsychological change in young people at high risk for schizophrenia: Results from the first two neuropsychological assessments of the Edinburgh High Risk Study. Psychological Medicine, 30, 1111–1121.

10.

Dohrenwend

B. P.

Levav

Shrout

P. E.

Schwartz

Naveh

Link

B. G.

… Stueve

(1992). Socioeconomic status and psychiatric disorders: The causation-selection issue. Science, 255, 946–952.

11.

Eaton

Harrison

(2001). Life chances, life planning, and schizophrenia: A review and interpretation of research on social deprivation. International Journal of Mental Health, 30, 58–81.

12.

Eaton

W. W.

(1985). Epidemiology of schizophrenia. Epidemiologic Reviews, 7, 105–126.

13.

Fearon

Kirkbride

J. B.

Morgan

Dazzan

Morgan

Lloyd

… Murray

R. M.

(2006). Incidence of schizophrenia and other psychoses in ethnic minority groups: Results from the MRC AESOP Study. Psychological Medicine, 36, 1541–1550.

14.

Goldberg

E. M.

Morrison

S. L.

(1963). Schizophrenia and social class. British Journal of Psychiatry, 108, 785–802.

15.

Goldberg

Fruchter

Davidson

Reichenberg

Yoffe

Weiser

(2011). The relationship between risk of hospitalisation for schizophrenia, SES, and cognitive functioning. Schizophrenia Bulletin, 37, 664–670.

16.

Hare

E. H.

Price

J. S.

Slater

(1972). Parental social class and psychiatric patients. British Journal of Psychiatry, 121, 515–524.

17.

Harrison

Gunnell

Glazebrook

Page

Kwiecinski

(2001). Association between schizophrenia and social inequality at birth: Case-control study. British Journal of Psychiatry, 179, 346–350.

18.

Heinrichs

R. W.

Zakzanis

K. K.

(1998). Neurocognitive deficit in schizophrenia: A quantitative review of the evidence. Neuropsychology, 12, 426–445.

19.

Hjern

Wicks

Dalman

(2004). Social adversity contributes to high morbidity in psychoses in immigrants- a national cohort study in two generations of Swedish residents. Psychological Medicine, 34, 1025–1033.

20.

Hollingshead

A. B.

Redlich

F. C.

(1958). Social class and mental illness: Community study. Hoboken, NJ: John Wiley & Sons, Inc.

21.

Jones

Rodgers

Murray

R. M.

Marmot

(1994). Child develop-mental risk factors for adult schizophrenia in the British 1946 birth cohort. Lancet, 344, 1398–1402.

22.

Kirkbride

J. B.

Barker

Yang

Jones

P. B.

Coid

J. W.

(2008). Psychoses, ethnicity and socio-economic status. British Journal of Psychiatry, 193, 18–24.

23.

Koenig

J. I.

Kirkpatrick

Lee

(2002). Glucocorticoid hormones and early brain development in schizophrenia. Neuropsychopharmacology, 27, 309–318.

24.

Liberatos

Link

B. G

Kelsey

J. L.

(1988). The measurement of social class in epidemiology. Epidemiologic Reviews, 10, 87–121.

25.

Mäkikyrö

Isohanni

Moring

Oja

Hakko

Jones

Rantakallio

(1997). Is a child’s risk of early onset schizophrenia increased in the highest social class? Schizophrenia Research, 23, 245–252.

26.

Mortensen

P. B.

Pedersen

C. B.

Westergaard

Wohlfahrt

Ewald

Mors

… Melbye

(1999). Effects of family history and place and season of birth on the risk of schizophrenia. New England Journal of Medicine, 340, 603–608.

27.

Mulvany

O’Callaghan

Takei

Byrne

Fearon

Larkin

(2001). Effect of social class at birth on risk and presentation of schizophrenia: Case-control study. BMJ, 323, 1398–1401.

28.

Timms

(1998). Gender, social mobility and psychiatric diagnoses. Social Science & Medicine, 46, 1235–1247.

29.

Weinberger

D. R.

(1987). Implications of normal brain development for the pathogenesis of schizophrenia. Archives of General Psychiatry, 44, 660–669.

30.

Wender

Rosenthal

Kety

Schulsinger

Welner

(1973). Social class and psychopathology in adoptees. Archives of General Psychiatry, 28, 318–325.

31.

Werner

Malaspina

Rabinowitz

(2007). Socioeconomic status at birth is associated with risk of schizophrenia: Population-based multilevel study. Schizophrenia Bulletin, 33, 1373–1378.

32.

Wicks

Hjern

Gunnell

Lewis

Dalman

(2005). Social adversity in childhood and the risk of developing psychosis: A national cohort study. American Journal of Psychiatry, 162, 1652–1657.

33.

Wiersma

Giel

De Jong

Slooff

C. J.

(1983). Social class and schizophrenia in a Dutch Cohort. Psychological Medicine, 13, 141–150.